Synthesis of the disaccharide moiety of bleomycin. 2-O-(3-O

Kiyoaki Katano, Alan Millar, Vince Pozsgay, John L. Primeau, and Sidney M. Hecht ... Kiyoaki Katano, Haoyun An, Yoshiaki Aoyagi, Mark Overhand, Steven...
0 downloads 0 Views 790KB Size
J. Org. Chem. 1986,51, 2927-2932 126 (ll),111 (32), 73 (34), 64 (39).

Desethyl- 15-thiono-20,21-didehydro-l6a-carbomethoxycleavamine Methiodide (51). A solution of 103mg (0.300 mmol) of the thionocleavamine 45b and 5 mL of methyl iodide in 10 mL of dichloromethane and 10 mL of benzene was stirred for 2 h,

2927

J = 6 Hz, 1 H), 7.15 (d, J = 6 Hz,1 H), 7.10 (t,J = 6 Hz,1H), 7.05 (t, J = 6 Hz,1 H),6.00 (dd, J = 5, 1 Hz,1 H), 4.42 (br d, J = 5 Hz, 1 H), 3.75 (s,3 H), 3.65-2.80 (m, 7 H), 2.65 (bra, 1H), 2.20 (s, 3 H), 1.85 (dd, J = 1 0 , l Hz,1H); direct inlet probe mass spectrum, m/z (relativeintensity) 354 (M', 38), 167 (12), 153 (30),

149 (25), 139 (17), 125 (33), 111(15), 86 (49), 84 (96), 69 (73), 57 resulting in precipitation of a red methiodide and complete re(89), 55 (100). The enol ether 49c structure could be tentatively action of the thione (TLC). Concentration, trituration with ether, and drying for 30 min at 0.1 mm provided 145 mg (99%) of the assigned by comparison of the NMR spectrum with that of the corresponding methoxy compound 49b (above). A reduction unstable product, mp 170-175 OC: TLC (SOz, methanol-triethylamiine) Rr 0.2 yellow, CAS, green; W (ethanol) A- 228,285, product structure 52 was tentatively assigned to the third product 293 nm; 270-MHz NMR (CDClJ 6 9.30 (br d, J = 3 Hz, 1 H), 9.05 on the basis of one olefin and one (C-16) NMR proton signal a t 6 5.65 and 5.2523and a mass spectrum with m / z 357 (M+ + 1, (br s, 1 H), 7.50-7.40 (2d, J = 6 Hz, 2 H), 7.20 (t,J = 6 Hz, 1H), 7.15 (t, J = 6 Hz, 1 H), 6.55 (d, J = 3 Hz, 1 H), 4.70 (m, 1 H), 100%) and 355 (M+ - 1,80%). 3.90 (d, J = 8 Hz, 1 H), 3.80 (m, 1 H), 3.70 (8, 3 H), 3.45 (m, 1 Desethylcatharanthine(50). Under argon, 300 mg of Raney H), 3.25-3.10 (m, 1 H), 3.70 (8, 3 H), 3.45 (m, 1H), 3.25-3.10 (m, nickel (50% in water at pH 10-11)was washed with 5 X 10 mL 3 H), 2.60 ( 8 , 3 H), 2.50 (m, 3 H); direct insertion probe mass of distilled water and with 4 X 10 mL of acetone and then heated spectrum (70 eV), m / z (relative intensity) 354 (M+- HI, 9), 322 at reflux for 2 h with 15 mL of acetone. The liquid phase was (5), 309 (13), 228 (4), 142 (loo), 127 (65). withdrawn and the solid catalyst washed with 5 X 10 mL of Deset hyl-15-methoxycatharanthine (49b). A solution of methanol and then suspended in 10 mL of methanol. Addition desethyl-15-thion~20,21-didehydr~l~-~bomethoxycleav~ine of 10 mg of the thioenol ether 49a in 1mL of methanol and heating methiodide (51, 20 mg, 0.04 mmol) and 0.5 mL of diisopropyla t reflux for 30 min resulted in complete conversion to deseethylamine in 4 mL of dry methanol was heated at reflux for 40 thylcatharanthine (50). A longer reaction time produced the min. Cooling, concentration under vacuum, and preparative TLC corresponding dihydro product desethylcoronaridine (mass on silica gel with ethyl acetate provided 9 mg of the enol ether spectrum, NMR). For desethylcatharanthine (50): no real mp 49b (70%) (R,0.2) and 1.5 mg (10%) of the thioenol ether 49a, could be observed under a microscope for a sample crystallized R, 0.3. For 49b: UV (ethanol) ,A, 230,285,293 nm; 270-MHz from ethyl acetate. Decomposition started around 160 "C (see NMR (CDC13) 6 7.70 (5, 1 H),7.50 (d, J = 6 Hz,1 H), 7.25 (d, note under dl-catharanthine). Reported mp 155-160 TLC J = 6 Hz, 1 H), 7.12 (t,J = 6 Hz, 1 H), 7.10 (t, J = 6 Hz, 1H), (SOz,diethyl ether-triethylamine) Rf 0.30; for the thioenol ether 5.25 (dd, J = 6, 2 Hz,1 H), 4.45 (br d, J = 6 Hz, 1 H), 3.85 (s, 49a R, 0.35; for desethylcoronaridine R, 0.47. Direct insertion 3 H), 3.55 (s, 3 H), 3.40-2.80 (m, 7 H), 2.65 (br s, 1H), 1.85 (dd, probe mass spectrum, m / z (relative intensity) 308 (M', 49), 229 J = 11, 2 Hz, 1 H); direct insertion probe mass spectrum, m / z (12), 214 (15), 170 (8), 168 (13), 167 (9), 154 (18), 124 (8), 108 (ll), (relative intensity) 338 (M', 60), 214 (9), 168 (21), 154 (18),137 107 (loo), 94 (36), 93 (28), 85 (21), 79 (23), 77 (7), 67 (20), 65 (7), (loo), 123 (37), 109 (55), 95 (16), 83 (23). For 49a, see below. 59 (8), 57 (7), 55 (10). The product gave a 270-MHz NMR Desethyl- 15-5-methylcatharanthine (49a), Desethyl-15spectrum which matched a 90-MHzspectrum provided by Prof. ethoxycatharanthine (49c), and Desethyl-l5,20-didehydroR. J. Sundberg. It had the same TLC R, value as a comparison 15-5-methylcleavamine (52). Repetition of the preceding resample from Virginia.33 action with 35 mg (0.073 mmol) of the methiodide in 8 mL of absolute ethanol and centrifugal chromatography of the product Acknowledgment. This work was supported by Grant on silica gel with ethyl acetate provided 17 mg (66%) of the R01-12010 from the National Cancer Institute of the Nathioenol ether 49a [TLC R, 0.2 (CAS, blue-green)] and small tional Institutes of Health. We thank Timothy Spitzer, amounts of the ethoxy compound 49c [Rf0.1 (CAS, violet)] and Patricia Matson, and Bruce Pitner of our group for mass the reduction product 52 [Rf 0.6 (CAS pink)]. For 49a: UV spectra. A valuable comparison sample of desethyl(ethanol) ,A, 230,285,293 nm; IR (film),,v 3383,3364,2945, catharanthine was kindly provided by Professor R. J. 2919,2877,2847,1728,1494,1460,1434,1343,1272,1255,1229, 1086, 744 cm-l; 270-MHz NMR (CDClJ 6 7.80 (s, 1H), 7.40 (d, Sundberg.

Synthesis of the Disaccharide Moiety of Bleomycin. 2- 0- (3- 0-Carbamoyl-a-~-mannopyranosy~)-~-gulopyranose Derivatives Kiyoaki Katano, Alan Millar, Vince Pozsgay, John L. Primeau, and Sidney M. Hecht*+ Departments of Chemistry and Biology, University of Virginia, Charlottesuille, Virginia 22901

Received January 14, 1986 The synthesis of the carbohydrate moiety of bleomycin [2-~-(3-~-carbamoy~-c~-~-mannopyranosy~)-~-gulopyranose] is described. A key parameter in defining a successful strategy was the lability of the carbamoyl group. Several approaches were investigated; the most successful involved the coupling of 1,6-di-O-acetyl-3,4-di-Obenzyl-&L-gulopyranoae(19) and 2 , 4 , 6 - t r i - 0 - a c e t y l - 3 - O - c a r b a m o y l - c ~ - ~ chloride ~ o p ~ ~ (17) ~ l via the agency of silver trifluoromethanesulfonate and tetramethylurea. Also reported is the synthesis of 1,6-di-O-acetyl-3,4di-O-benzyl-2-O-(2,3,4,6-tetra-~-acetyl-c~-~-mannopyranosyl)-~-gulopyranose (16), a dissacharide useful for the synthetic elaboration of decarbamoyl bleomycin.

Our continuing interest in the synthesis of bleomycin group antibiotics' necessitated the synthesis of the carbohydrate moiety of bleomycin [2-0-(3-O-carbamoyl-a-Dmannopyranosy1)-L-gulopyranose(111 on a preparative 'University QfVirginia and Smith Kline & French Laboratories. To whom correspondence should be addressed at the Department of Chemistry, University of Virginia.

scale and in a form suitable for further elaboration. The successful synthesis of llb required access to suitably blocked derivatives of L-gulose2 and to activated 3-0(1) (a) Minster, D. K.; Hecht, S. M.J. Org. Chem. 1978,43,3987. (b) pozsgay, v.; Oh& T.;He&, S.M. J . Org. Chem. 1981, 46, 3761. (c) Aoyagi, Y.;Katano, K.; Supune,H.; Primeau,J. L.; Chmg, L.-H.; Hecht,

S. M: J . Am. Chem. Sac. 1982, 104, 5537.

0022-326318611951-2927$01.50/00 1986 American Chemical Society

2928 J. Org. Chem., Vol. 51, No. 15, 1986

Katano et al.

w::

Ho*HO oH

7

6 Ho 0% "2

1

carbamoyl-D-mannose derivative^,^ the preparations of which have been reported. Herein we describe the synthesis of the disaccharide moiety of bleomycin as well as several protected derivatives potentially useful for further elaboration to bleomycin and bleomycin congeners that can help to define the mechanism of action of this group of antibiotics.

8

BnO

Results and Discussion a-

Incorporation of the dissaccharide moiety into the bleomycin molecule involves the regio- and stereospecific incorporation of L.-erythro-/3-hydroxyhistidineat 0-1of the L-gulose moiety. The configuration required at 0-1is a, Le., involving 1,2-cis glyc~sylation.~Because a nonparticipating protecting group would logically be required at 0-2 to permit access to the requisite configuration at 0-1, either of two synthetic strategies seemed possible. The first of these (Scheme I, route a) would provide an L-gulose derivative, e.g., 2,' containing a nonparticipating alkyl group on 0-2. Thus glycosylation with an appropriately blocked P-hydroxyhistidine derivative would be expected to give a 1,2-cis glycoside (3); deprotection at 0 - 2 and condensation with an appropriate mannopyranosyl halide3 could then be expected to provide 5. Alternatively (route b), initial deprotection at 0 - 2 and condensation with the mannosyl halide would provide a disaccharide (4) containing a mannose group at 0 - 2 of L-gulose. Glycosylation of the disaccharide at 0-1of the L-gulose moiety would again be expected to provide the requisite 1,2-cis glycoside (5). Initial investigation of these two routes quickly established that route b was refer red.^ Coupling of the readily available (acetylcarbamoy1)mannosyl bromide 6 with anhydro-L-gulose derivative 7 (AgOTf, (CH3)2NCON(CH3)2)6 gave disaccharide 8 in 68% yield. That the mannose glycosyl bond had the a-configuration was indicated by the magnitude of the coupling constant observed for the anomeric proton of mannose (6 4.89, Jl,z-1 Hz).While further synthetic elaboration of the disaccharide clearly required solvolysis of the 1,6anhydro linkage, no difficulty was envisaged in view of the relatively mild procedures available for such transformat i o n ~ . ~In ~ ~fact, treatment of disaccharide 8 with Ac20/HOAc (0 "C, catalytic H,SO,) provided 9 in 98% yield as a 1 5 mixture of a and (3 anomers, respectively.

(2) Katano, N.; Chang, P.-I.;Millar, A.; Pozsgay, V.; Minster, D. K.; Ohgi, T.; Hecht, S. M. J. Org. Chem. 1985,50, 5807. (3) Millar, A.; Kim, K. H.; Minster, D. K.; Ohgi, T.; Hecht, S. M. J. Org. Chem. 1986,51, 189. (4) Bochkov, A. F.; Zaikov, G. E. In Chemistry of the 0-Glycosidic Bond: Formation and Cleauage; Pergamon Press: Oxford, 1979. (5) Investigation of route a indicated that the conversion 2 3 proceeded satisfactorily but that further conversion to 5 did not. (6) Hanesaian,S.;Banoub, J. Methods Carbohydr. Chem. 1980,8,247. (7) Cernag, M.; Stanik, J. Adv. Carbohydr. Chem. Biochem. 1977,34, 23 and references therein.

-

AcO

/

NHAc 9

Alternatively, hydrogenation of 8 over 10% palladium on carbon provided debenzylated disaccharide loa. Peracetylated disaccharide 11 was then obtained via acetylation (loa lob) and acetolysis of the 1,6-anhydro linkage. The ratio of a$ anomers was 1 5 , as indicated by 'H NMR (6 5.88, Jl,z= 8.4 Hz, (3 anomer; 6 6.28, J1,2 = 4.4 Hz,a anomer).

-

-

10a 10b

R=H

R=Ac

"O*OAC Ac 0

+y=$z:;. O